Publication date September 11th, prior to Redfield's testimony before Congress.
He committed perjury and there is proof. His own agency's report documents that masks have no statistical benefit.
Note that unlike the other data in this report this table has no CI bands displayed. The reason is obvious when you look at it; the two associations that are statistically-significant are being in close contact with a known case (duh) and having a family member who was the close case. Interestingly enough what is also borne out in the data is that people who are friends are negatively associated (in other words you're probably not getting it from someone asymptomatically as your friends stay away from you when they know they're sick) and the same negative association is present for work colleagues.
But the screaming data is at the bottom. 70.6% of the people who got the bug always wore cloth face coverings (or better) in the 14 days prior, .vs. only 74.2% who did not get the bug, and there was only one tick difference between those who "often" wore a mask.
This is statistical non-confirmation folks and while this is an observational study and thus cannot prove causation observational studies can disprove a thesis. In this case exactly that occurred: It disproved the effectiveness of masks.
This is in fact the CDC's own weekly report folks and it was issued five days before Redfield testified before the Senate.
He LIED, knowingly. His own agency produced and published this report five days before his testimony.
Lock that SOB up for perjury and drop all mask mandates. They're worthless based on the CDC's own statistical data.